Zen GGUF models

83 GGUF models in the Zen family, quantized by 10 uploaders. Sizes run from 33M to 79.7B parameters. The longest trained context is 256K tokens.

TinySmallMediumLarge
Most downloadedNewestLargestLongest contextSmallest
ZE
Zen Voyager
Zen · mradermacher
Params32.8B
Context40K
Q4_K_M19.8 GB
GGUF
ZE
Zen4 Coder
Zen · mradermacher
Params79.7B
Context256K
Q4_K_M48.5 GB
Coding
ZE
Zen3 Vl
Zen · mradermacher
Params8.2B
Context256K
Q4_K_M5.0 GB
Vision
ZE
zen 4b i1
Zen · mradermacher
Params4.02B
Context32K
Q4_K_M2.5 GB
ReasoningTool use
ZE
Zen Vl 8b Instruct
Zen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
Vision
ZE
xysera zen 4b
Zen · bartowski
Params4.02B
Context32K
Q4_K_M2.6 GB
ReasoningTool use
ZE
Zen4 Thinking
Zen · mradermacher
Params79.7B
Context256K
Q4_K_M48.5 GB
Reasoning
ZE
Zen Pro
Zen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
ZE
Zen Vl 8b Instruct
Zen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
Tool useVision
ZE
Zen3 Nano
Zen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
ZE
Zen Vl 8b Agent
Zen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
Tool useVision
ZE
Zen3 Vl
Zen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
Tool useVision
ZE
Zen4 Nano
Zen · mradermacher
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool use
ZE
Zen Vl 4b Instruct
Zen · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
Tool useVision
ZE
Zen4 Nano
Zen · mradermacher
Params752M
Context256K
Q4_K_M0.5 GB
ReasoningTool useVision
ZE
Zen4 Coder Pro
Zen · mradermacher
Params79.7B
Context256K
Q4_K_M48.5 GB
CodingTool use
ZE
Zen Guard Stream 4b
Zen · mradermacher
Params3.09B
Context32K
Q4_K_M1.9 GB
Tool use
ZE
zen 4b
Zen · mradermacher
Params4.02B
Context32K
Q4_K_M2.5 GB
ReasoningTool use
ZE
Zen Guard Gen 8b
Zen · mradermacher
Params7.62B
Context32K
Q4_K_M4.7 GB
Tool use
ZE
zen4 i1
Zen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
ZE
Zen Vl 30b Instruct
Zen · mradermacher
Params30.5B
Context256K
Q4_K_M18.6 GB
Tool useVision
ZE
Zen4 Thinking
Zen · mradermacher
Params79.7B
Context256K
Q4_K_M48.5 GB
ReasoningTool use
ZE
Zen Vl 4b Agent
Zen · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
Tool useVision
ZE
Zen Eco 4b Instruct
Zen · mradermacher
Params3.09B
Context32K
Q4_K_M1.9 GB
Tool use
ZE
Zen Vl 8b Agent
Zen · mradermacher
Params8.19B
Context256K
Q4_K_M5.0 GB
Tool useVision
ZE
Zen Pro
Zen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
ZE
Zen Vl 4b Instruct
Zen · mradermacher
Params4.02B
Context256K
Q4_K_M2.5 GB
Tool useVision
ZE
Zen Vl 30b Instruct
Zen · mradermacher
Params30.5B
Context256K
Q4_K_M18.6 GB
Tool useVision
ZE
zen4
Zen · mradermacher
Params8.19B
Context40K
Q4_K_M5.0 GB
ReasoningTool use
ZE
Zen Nano 0.6b
Zen · mradermacher
Params752M
Context40K
Q4_K_M0.5 GB
ReasoningTool use

About the Zen GGUF models

Zen is represented on ggufmodels.com by 83 GGUF models, quantized by 10 uploaders on Hugging Face. The most downloaded are Zen Voyager, Zen4 Coder and Zen3 Vl. Parameter counts run from 33M to 79.7B, and the longest trained context window is 256K tokens.

A Q4_K_M file of a Zen model is between 0.0 to 48.5 GB, which is the quant most people start with. Most files come from mradermacher (65), keisuke-miyako (6) and RockMan256 (4). The quants you will meet most often are Q4_K_M (77), Q5_K_M (71), Q6_K (68) and Q3_K_M (67), where the number is how many models offer that quant.

Frequently asked questions

Which Zen GGUF should I download?

The most downloaded Zen GGUF here is Zen Voyager. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.

How much VRAM does a Zen model need?

Q4_K_M files of Zen models on this site range from 0.0 to 48.5 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.

Where do Zen GGUF files come from?

The most common quantizers here are mradermacher, keisuke-miyako and RockMan256. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.

Keep browsing